SMUC '10 Proceedings of the 2nd international workshop on Search and mining user-generated contents
| Hi-index | 0.00 |
It is our great pleasure to welcome you to the 1st International CIKM Workshop on Topic-Sentiment Analysis for Mass Opinion Measurement -- TSA'09. This workshop seeks to bring together researchers in both computer science and social sciences who are interested in developing and using topic-sentiment analysis methods to measure mass opinion, and to foster communications between the research community and industry practitioners as well. The call for papers attracted 21 submissions from 14 countries. The program committee accepted 12 papers that cover a variety of topics, including topic-sentiment modeling, sentiment classification and retrieval, sentiment corpus construction, and applications of topic-sentiment analysis in text summarization, question answering, and recommender systems. The proposed approaches analyze opinions at all levels of granularity: clause, sub-sentence, sentence, paragraph and document. Most of the approaches combine machine learning and statistical methods and the use of linguistic resources (sentiment lexicons, syntactic rules, etc.) for sentiment identification. User-generated content (UGC) is still the main source of data for topic-sentiment analysis in various domains, like customer reviews, blogs, and discussion boards. In addition to English, corpora of other languages (Chinese, Spanish, and Portuguese) have also been studied. Many authors chose to manually annotate their own sentiment corpora to train machine learning algorithms, or employed automatic methods to acquire the sentiment annotation. This indicates the strong demand for large volume of annotated data in various topics and domains to facilitate topic-sentiment analysis.